Filter 超集将时间筛选器从仪表板传递到自定义查询

在一个切片中，需要自定义查询来合并来自不同筛选器的2个数据集。 每个数据集都有其日期/时间文件管理器。超集如何知道要用于哪个查询的日期名

下面是自定义查询的示例

1 SELECT "PRE" AS "Stage","Tutorial Flight"."Travel Date" AS "Travel_Date",
2       "Tutorial Flight"."Airline" AS "Airline",
3       sum("Cost") AS "Cost"
4 FROM  "Tutorial Flight"
5 WHERE 
6 "Travel Date" = strftime('%Y-%m-%d %H:%M:%S.000000','{{ from_dttm.isoformat() }}')
7 AND  "Tutorial Flight"."Airline"  in ( {{ "'" + "','".join(filter_values('Airline')) + "'" }} )
8 GROUP BY 
9        "Travel Date", "Airline"
10 UNION
11 SELECT "POST" AS "Stage","Tutorial Flight"."Travel Date" AS "Travel_Date",
12       "Tutorial Flight"."Airline" AS "Airline",
13       sum("Cost") AS "Cost"
14 FROM "Tutorial Flight"
15 WHERE 
16 "Travel Date" = strftime('%Y-%m-%d %H:%M:%S.000000','{{ from_dttm.isoformat() }}')
17 AND  "Tutorial Flight"."Airline"  in ( {{ "'" + "','".join(filter_values('Airline')) + "'" }} )
18 GROUP BY "Travel Date", "Airline"

本例中使用了3个过滤器。 日期\日期前的时间 日期\发布日期和时间 适用于这两个查询的航空公司

我们如何在from_dttm.isoformat（）中定义以使用表中的正确日期时间 1.来自同一个表的日期筛选器 2.来自不同表的日期筛选器

我试过了，但它总是只用一张桌子。 下面是来自一次性过滤器的json

{"cache_key": null, "cached_dttm": null, "cache_timeout": 86400, "error": null, "form_data": {"datasource": "18__table", "viz_type": "filter_box", "slice_id": 295, "url_params": {}, "granularity_sqla": "Travel Date", "time_range": "2020-06-15T00:00:00 : 2020-06-15T00:00:00", "filter_configs": [], "date_filter": true, "instant_filtering": true, "show_sqla_time_granularity": false, "show_sqla_time_column": false, "show_druid_time_granularity": false, "show_druid_time_origin": false, "adhoc_filters": [], "where": "", "having": "", "having_filters": [], "filters": []}, "is_cached": false, "query": "", "status": null, "stacktrace": null, "rowcount": 0, "data": {}}

请告知。

---

非常感谢。

您可以传递

{{from_dttm}

和

{to_dttm}

您可以传递

{{from_dttm}

和

{to_dttm}